Event Stream Overview

Export granular data for independent analysis with Connect. Gain insights about your audience and how users respond to marketing messaging .

Available Data Sets

The following Connect data sets documentation show all the information provided by each data set. Each bullet point in the following lists are linked to the full table.

Event Stream updates are exported hourly.

Email Event Data

The following schema reflects the events that can occur when a marketer sends an email through the platform, and the end user's interaction with that message.
  • Campaign Sends
  • Campaign Opens
  • Campaign Clicks
  • Campaign Bounces
  • Campaign Unsubscribes
  • Triggered Sends
  • Triggered Opens
  • Triggered Clicks
  • Triggered Bounces
  • Triggered Unsubscribes

Lookup Data

The following schema reflects the associated metadata that a marketer may wish to enrich event data for the purposes of segmentation in their reporting.

  • Client
  • Campaign
  • Template
  • LO Flow
  • Campaign A/B Test
  • Label

Mobile Event Data

The following schema reflects the events that can occur when a marketer sends a mobile message or push through the platform, and the end user's interaction with the message or the app itself.

  • Notifications
  • Device Events
  • Sessions
  • Custom Events
  • Impressions

Forward Compatibility

Engage by Sailthru will continue to iterate on its Event Stream offering in an additive manner (sometimes referred to as "forward compatibility"). New fields that appear in the file outputs will not necessarily match older schemas, and will generally be forward compatible. Engage by Sailthru advises that customer teams refer to a field by column name, and not by position. We recommend that customers avoid expecting or validating against a fixed number of fields, or fields in certain positions, which would cause errors if additional columns are provided or if field ordering changes. Sailthru will look to give advance warning in the case of any changes, and will work to minimize breaking changes to the schema at any time (i.e. field name changes, field removals).

Bucket Structure

Your bucket structure will look different depending on the type of event/lookup data. Data exported into your bucket will adhere to the following pattern for event data tables: 

s3://[bucketname]/[table_name]/dt=[yyyymmdd]/client_id=[client_id]/[exportfilename].csv
For example: 
s3://example-bucket/campaign_click/dt=20190926/client_id=0000/part-1ae5f73f.csv
Data exported into your bucket will adhere to the following pattern for lookup data tables: 
s3://[bucketname]/[table_name]/dt=[yyyymmdd]/hr=[hh]client_id=[client_id]/[exportfilename].csv
For example: 
s3://example-bucket/campaign/dt=20191001/hr=00/client_id=0000/part-1ae5f73f.csv

Note: Engage by Sailthru recommends that you avoid adding your own objects into this bucket. The intention is to avoid unexpected issues with the export process or inadvertently giving Sailthru access to non-Sailthru data.

Bucket Policy

To set up your bucket policy, see S3 Bucket Policy Information. This page contains information on setting up an Event Stream bucket policy, a joint policy for both Event Stream and Data Exporter, or a Data Exporter connection.

Event Stream Errors

What happens if an Event Stream export fails? We classify errors into two categories, General Error or Client Service Error.

  • General Error - are any errors not caused by the client's setup. Engage by Sailthru will still export data for the entire day and it will be picked up in the next planned export in the same day. If there is not another export planned for the same day, a re-run can be ordered.
  • Client Service Error - before triggering the export, Engage by Sailthru checks to ensure the client's S3 bucket is accessible for get, put, and delete operations. If this fails, client is excluded from that export.

How are you notified of a failure?

  • If failures are resolved fairly quickly on their own, then it will be picked up in the next planned export.
  • If there are widespread and persistent failures occurring then this will trigger our incident process and you will be notified.
  • If you identify an issue with your exports, contact the Support team for assistance in identifying what caused the failure, General or Client Service Error, to determine the appropriate resolution.

Is there any limit on failed catchup? Example: if Event Stream fails all day long, at the next successful export, is there a limit to how much data would be sent?

  • No, there is no limit.